Systems and methods for enabling conversational access to tabular data
Methods, non-transitory computer readable media, and a data server that assist with enabling conversational access to tabular data includes determining in response to a user input, tabular data comprising a header row with header data in each column and one or more table data rows with row data in each column. A first prompt is provided to a large language model to generate a dummy table comprising the header row and a dummy row with dummy row data in each column and a dummy table is received. An alias table comprising the header row and an alias row with alias row data in each column is generated. A second prompt is provided to the large language model to generate a dummy row text representation of the dummy row data, wherein the dummy row text representation includes the alias row data inserted as placeholders of the dummy row data and the dummy row text representation is received. A row text representation is generated for each of the one or more table data rows by replacing the alias row data in the dummy row text representation with row data of corresponding ones of the one or more table data rows.
1 . A method comprising:
determining, by a data server, in response to a user input, tabular data comprising a header row with header data in each column and one or more table data rows with row data in each column;
providing, by the data server, a first prompt to a large language model to generate a dummy table comprising the header row and a dummy row with dummy row data in each column;
receiving, by the data server, the dummy table;
generating, by the data server, an alias table comprising the header row and an alias row with alias row data in each column;
providing, by the data server, a second prompt comprising header data associated with the dummy row data and the header data associated with the alias row data, to the large language model to generate a dummy row text representation of the dummy row data, wherein the dummy row text representation includes the alias row data inserted as placeholders of the dummy row data;
receiving, by the data server, the dummy row text representation;
determining, by the data server, a correlation between the row data of the one or more table data rows and the alias row data by replacing in the tabular data, the header data of each column with the corresponding alias row data determined from the alias table;
generating, by the data server, a row text representation for each of the one or more table data rows by replacing the alias row data in the dummy row text representation with row data of corresponding ones of the one or more table data rows based on the correlation;
receiving, by the data server, a user query from the user device;
semantically comparing, by the data server, the user query to one or more of the generated row text representations to determine an answer to the user query; and
providing, by the data server, to the user device, the determined answer as a response to the user query.
2 . The method of claim 1 , wherein the tabular data is stored as a database table.
3 . The method of claim 1 , wherein each row text representation is a semantic summary of one of the one or more table data rows.
4 . A data server comprising:
one or more processors; and
a memory coupled to the one or more processors which are configured to execute programmed instructions stored in the memory to:
determine, in response to a user input, tabular data comprising a header row with header data in each column and one or more table data rows with row data in each column;
provide a first prompt to a large language model to generate a dummy table comprising the header row and a dummy row with dummy row data in each column;
receive the dummy table;
generate an alias table comprising the header row and an alias row with alias row data in each column;
provide a second prompt comprising header data associated with the dummy row data and the header data associated with the alias row data, to the large language model to generate a dummy row text representation of the dummy row data, wherein the dummy row text representation includes the alias row data inserted as placeholders of the dummy row data;
receive the dummy row text representation;
determine a correlation between the row data of the one or more table data rows and the alias row data by replacing in the tabular data, the header data of each column with the corresponding alias row data determined from the alias table;
generate a row text representation for each of the one or more table data rows by replacing the alias row data in the dummy row text representation with row data of corresponding ones of the one or more table data rows based on the correlation;
receive a user query from the user device;
semantically compare the user query to one or more of the generated row text representations to determine an answer to the user query; and
provide to the user device the determined answer as a response to the user query.
5 . The data server of claim 4 , wherein the tabular data is stored as a database table.
6 . The data server of claim 4 , wherein each row text representation is a semantic summary of one of the one or more table data rows.
7 . A non-transitory computer readable medium storing instructions which when executed by one or more processors, causes the one or more processors to:
determine, in response to a user input, tabular data comprising a header row with header data in each column and one or more table data rows with row data in each column;
provide a first prompt to a large language model to generate a dummy table comprising the header row and a dummy row with dummy row data in each column;
receive the dummy table;
generate an alias table comprising the header row and an alias row with alias row data in each column;
provide a second prompt comprising header data associated with the dummy row data and the header data associated with the alias row data, to the large language model to generate a dummy row text representation of the dummy row data, wherein the dummy row text representation includes the alias row data inserted as placeholders of the dummy row data;
receive the dummy row text representation;
determine a correlation between the row data of the one or more table data rows and the alias row data by replacing in the tabular data, the header data of each column with the corresponding alias row data determined from the alias table;
generate a row text representation for each of the one or more table data rows by replacing the alias row data in the dummy row text representation with row data of corresponding ones of the one or more table data rows based on the correlation;
receive a user query from the user device;
semantically compare the user query to one or more of the generated row text representations to determine an answer to the user query; and
provide to the user device the determined answer as a response to the user query.
8 . The non-transitory computer readable medium of claim 7 , wherein the tabular data is stored as a database table.
9 . The non-transitory computer readable medium of claim 7 , wherein each row text representation is a semantic summary of one of the one or more table data rows.